Advanced methodologies for selection and recommendation of genotypes across diverse environments

Abstract

The evaluation and selection of genotypes in multi-environment trials (MET) is challenging due to the large volume and complexity of data, especially when considering genotype-by-environment (G×E) interactions. To contribute to this field, two studies were conducted exploring methodologies and applications aimed at genotype selection in MET experiments. The first chapter focuses on eucalyptus, utilizing factor analytic mixed models (FAMM) to assess the stability and adaptability of clones across different environments in Brazil. The research was based on data from clonal trials conducted in 11 experiments across distinct regions, proposing the application of genotype-ideotype distance for selecting high-performance clones. A comparison with the OP-RMSD method was also conducted. Additionally, the application of different factor rotations (varimax and SVD) is discussed, aiming to facilitate the interpretation of G×E interactions, simplify environmental stratification, and enable the identification of clones with broad adaptability. The second chapter addresses potato breeding, based on data from the National Chip Processing Trial (NCPT), aiming to identify highly productive and stable clones in different regions of Florida. This study integrated phenotypic data, molecular markers, and environmental covariates, analyzed using FAMM and methods such as genotype-ideotype distance and overall performance (OP) combined with root mean square deviation (RMSD). The use of Geographic Information Systems (GIS) allowed for spatial visualization of the results, contributing to practical recommendations for the potato chip market. Both studies highlight the importance of integrating genotypic, phenotypic, and environmental data with advanced statistical tools, offering solutions to overcome the challenges posed by environmental variability. Furthermore, they emphasize the need to improve crop productivity and stability. The methodology described in both chapters is versatile, and adaptable to other breeding programs, positioning itself as a robust approach to meet the regional and global demands of modern agriculture.